Query Insights search now supports negated filter terms for both Postgres and Vitess. Prefix any search term with ! to exclude matching results — for example, !table:users returns all queries that don't touch the users table, and !indexed:true surfaces unindexed queries. Positive and negative terms can be combined freely: p99:>100 !table:sessions finds slow queries that don't involve the sessions table. Negation applies to the token immediately following !, so quoted phrases work too — !"some text" excludes queries matching that phrase, while "!some text" treats the exclamation mark as a literal character when searching for queries that include it in the normalized SQL. Read more about filtering queries with Vitess or filtering queries with Postgres.
Full Changelog: https://github.com/planetscale/planetscale-go/compare/v0.162.0...v0.163.0
We have added a new GCP region. Now you can deploy Postgres and Vitess databases on GCP in Eemshaven, Netherlands (gcp-europe-west4). For more information about supported regions, see our regions documentation.
You can now configure storage when you create a PlanetScale Postgres database. For Postgres databases using Amazon Elastic Block Storage (EBS), you can now set storage size during database creation. This is available in the web dashboard, the API, and the CLI. API: Create databaseCLI: pscale database create
Full Changelog: https://github.com/planetscale/planetscale-go/compare/v0.161.0...v0.162.0
Full Changelog: https://github.com/planetscale/planetscale-go/compare/v0.160.0...v0.161.0
We've added a new API endpoint that lets you check whether a deploy request's target branch has enough storage to safely deploy schema changes. The endpoint returns details about storage capacity across keyspaces and shards, including how much additional storage is needed and whether the cluster can be upgraded for more space. This is useful for catching potential storage issues before deploying large schema changes on Vitess databases. Read more
A new variant of the PlanetScale MCP server is now available that provides access to Insights data and Schema Recommendations without exposing query execution tools. Use this server when you want AI agents to analyze query performance and suggest optimizations but don't need them to run queries directly against your database. Server URL:https://mcp.pscale.dev/mcp/planetscale-insights-only
This is ideal for workflows like the self-improving database, where agents review Insights data and open pull requests with performance improvements. Learn more about the PlanetScale MCP server
A new variant of the PlanetScale MCP server is now available that provides access to Insights data and Schema Recommendations without exposing query execution tools.
A new API endpoint has been added that lets you check whether a deploy request's target branch has enough storage to safely deploy schema changes.
RELEASE_GITHUB_TOKEN with a GitHub App tokenwarning_thresholdscript/release.sh is idempotentskipFull Changelog: https://github.com/planetscale/planetscale-go/compare/v0.159.0...v0.160.0
Full Changelog: https://github.com/planetscale/planetscale-go/compare/v0.158.0...v0.159.0
PlanetScale for Postgres now supports the vectorscale extension. vectorscale is a PostgreSQL extension for high-performance approximate nearest neighbor (ANN) search on vector embeddings. You can enable vectorscale from the Extensions tab on your Clusters page. Learn more about Postgres extensions
PlanetScale for Postgres now supports the vectorscale extension.
RELEASE_VERSIONplanetscale-actions-botlint.yml